Currently, there are various systems that can mix communication streams such as video and audio streams in a peer-to-peer environment. In existing peer-to-peer systems, users can set up three- or multi-party conference calls from their communication device. The communication node that the originator used to setup the conference is defined as the mixing node. These types of systems have problems when the originator of the conference wants to drop out of the conference while the other participants want to stay on the conference. When the originator of the conference exits, the conference is dropped. The other participants must make a new communication in order to continue.
Other systems, such as the system described in U.S. Patent Publication No. 2008/0177833, disclose a peer-to-peer system that creates an initial overlay mesh network that identifies an initial mixing node. Later, hierarchical networks of mixers are created to stream the conference in a more efficient manner. The problem with this system is that nodes in the network constantly have to send messages to each other even if a conference is not established. Moreover, a large mixing hierarchy may introduce latency. This creates unnecessary overhead and delay in the network.